30 Aug 2012

Never mind assay reliability, protocol standardization, and other technicalities of translational science; the hunt for blood-based Alzheimer’s biomarkers has struggled with a more fundamental requirement of scientific research—that of reproducibility of the basic findings. In this regard, new studies have taken a step forward. Several research groups have cross-validated plasma markers in separate U.S. and Australian cohorts, and in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset, and some of the common analytes seem to associate with AD markers in cerebrospinal fluid (CSF). The ultimate goal—a blood test to screen for Alzheimer’s—remains a way off, but the recent work should invigorate the quest, and highlights key challenges the field must overcome.

Reporting in the August 28 Neurology, scientists identified four plasma proteins that associated with AD or mild cognitive impairment (MCI) in 600 subjects from two separate cohorts. The four were later confirmed in ADNI samples. In similar fashion, other groups found AD-associated proteins in plasma samples from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study and Texas Alzheimer’s Research Consortium (TARC), and validated some of these analytes in the ADNI dataset.

In the search for AD biomarkers, scientists have focused heavily on CSF and neuroimaging measures, with plasma markers receiving comparatively scant attention. Yet a blood test would be simpler and cheap, potentially becoming a widespread screening tool, whereas spinal taps and brain scans are more invasive or expensive, respectively, which in the U.S. has so far largely limited these procedures to research cohorts, said Lenore Launer, who heads the neuroepidemiology unit at the National Institute on Aging in Bethesda, Maryland. “Any translation of what we find in research populations into a community setting is going to be more difficult if we don’t have plasma markers.”

For the Neurology study, first author William Hu of Emory University School of Medicine, Atlanta, Georgia, and colleagues asked a simple question. “Before we even look at diagnostic accuracy, can we replicate the results across centers in a fairly sizeable population?” asked Hu. The researchers measured levels of 190 plasma proteins in 600 subjects enrolled in research at the University of Pennsylvania School of Medicine, Philadelphia, and at Washington University School of Medicine, St. Louis, Missouri. The UPenn cohort included 126 seniors with normal cognition, 16 MCI patients, 88 AD patients, and 37 people with other dementias. The WashU group had 242 controls with a Clinical Dementia Rating (CDR) score of 0, 63 with CDR of 0.5, and 28 with CDR of 1.

Among the 190 proteins assayed, 23 varied between controls and impaired subjects in both cohorts, though only 17 were associated in the same direction. These 17 were analyzed in the ADNI cohort (58 cognitively healthy, 396 MCI, 112 AD) for validation. Four were found highly associated with AD/MCI: ApoE, B-type natriuretic peptide (BNP), C-reactive protein (CRP), and pancreatic polypeptide. What’s more, levels of these four proteins associated with pathological AD measures of Aβ42 and tau/Aβ42 in the CSF. That is important, because plasma Aβ never correlated well with CSF Aβ/tau.

“These findings are important because they demonstrate that consistent blood-based markers can be identified across independent cohorts, and that these markers are related to CSF markers,” Sid O’Bryant, University of North Texas Health Sciences Center, Fort Worth, wrote in an accompanying editorial in Neurology.

Ralph Martins of the University of Melbourne, Australia, was senior investigator of the AIBL study, which was published online July 16 in the Archives of Neurology. James Doecke of the Australian E-Health Research Centre, Herston, and Simon Laws of Edith Cowan University, Joondalup, were first authors. Their team measured 151 plasma proteins in an even larger sample—754 healthy elderly and 207 AD patients from the Australian cohort—then validated the findings in the ADNI dataset. This analysis yielded 18 hits, including two from the Penn/WashU study—ApoE and pancreatic polypeptide.

As reported in PLoS One in April, O’Bryant and colleagues analyzed plasma samples from about 400 controls and AD patients from the Texas Alzheimer’s Research Consortium (TARC). They identified 11 proteins that associated with AD status in TARC and were then confirmed in the ADNI cohort (O’Bryant et al., 2011). The shared list includes CRP and pancreatic polypeptide.

In other recent analyses, researchers found that plasma markers from the ADNI cohort associate with apolipoprotein E genotype (Soares et al., 2012), and that longitudinal data can improve the accuracy of AD-associated blood markers (Johnstone et al., 2012).

Broadly speaking, the studies suggest that “there does appear to be a blood-based signature associated with all stages of AD,” Holly Soares, Bristol-Myers Squibb, told Alzforum. However, she and others note that the field remains nascent and continues to wrestle with methodological and conceptual issues.

Perhaps the biggest advantage of plasma markers is the ease of drawing blood—but this also presents a challenge. “Anybody can [draw blood], so there are many protocols for doing it,” Hu said. “No two phlebotomists draw blood the same way.” Even within research sites, some keep the tourniquet on during the whole draw, whereas others take it off partway through; not everyone uses the preferred vein first; and draw volumes differ depending on whether they are done strictly for research or as part of a series of blood samples collected for clinical purposes. “How the blood is drawn and how samples are processed can make varying degrees of differences according to the analyte you are looking at,” Hu said.

Moreover, individual analytes have quirks. Take CRP, for example, a protein found to be associated with AD/MCI in the WashU/UPenn and TARC studies. CRP levels vary throughout the day, peaking during early morning hours (Koc et al., 2010), which might skew results if blood draws are taken at different times for different subjects in a study.

A Neurology paper posted online August 15 also illustrates how factors such as age and family history can complicate analysis of a particular protein. High levels of CRP raise a person’s risk for heart disease, and some studies have linked elevated CRP to increased risk for cognitive decline and Alzheimer’s disease (e.g., Schmidt et al., 2002; Marioni et al., 2009). Yet, in people over 75, the association between CRP and cognition has been less consistent, sometimes even opposite in direction. In the present study, Jeremy Silverman of Mount Sinai School of Medicine, New York, and colleagues took things further and found an interesting twist. They looked at men 75 and older who stayed mentally intact despite their high serum CRP, and asked whether they had inherited some form of natural protection. To test that idea, they looked to the subjects’ extended families and found that the relatives of seniors with high CRP were less likely to develop dementia than are relatives of seniors with lower CRP. The findings suggest to the authors that elderly with elevated CRP may have protective factor(s) that guard against dementia, hinting at a complex relationship between CRP and AD.

Another consideration in the recent plasma analyses is that cross-validation was done in the ADNI dataset, though some researchers noted ADNI controls are few in number (fewer than 60) and may not reflect the general population. “ADNI controls were carefully selected based on high CSF Aβ42 levels. This means they had a lower frequency of ApoE4 genotype than the AD or MCI comparison groups, and low for a control population in general, which makes it difficult to generalize these results,” said Douglas Galasko of the University of California, San Diego. “For validation of a plasma signature in diagnosis, samples should also be obtained from patients with other neurological disorders (e.g., Parkinson’s disease) and from a less select group of controls.” Along those lines, Nicole Schupf of Columbia University, New York, considered it a strength that the WashU subjects in the study by Hu et al. were classified by CDR scores instead of disease, suggesting that people with non-AD dementias were likely included in the analysis.

Besides these technical issues, a conceptual question dogs plasma analyses, as well. WashU’s Anne Fagan, a coauthor on the Neurology paper, said, “It may be easier to defend the idea that CSF reflects what goes on in the brain better than plasma. It bathes the brain, and many molecules diffuse readily from CSF to interstitial fluid. Even if we had the best assays to measure an analyte in plasma and standardization was achieved, does it reflect the true biology of the brain?” Furthermore, Galasko noted, “There could be a systemic response to having AD that leads to a certain substance going up or down in plasma. Is that an AD-specific response?”

Plasma measurements are tricky because there are considerably more proteins—and hence, more factors that degrade them—in the blood than in CSF. This means that even if the blood does reflect what is happening in the brain, “you may have to catch it immediately upon entry into the peripheral compartment,” Fagan said.

And, of course, there remains the challenge of replicating findings. In a widely noted 2007 Nature Medicine paper, Tony Wyss-Coray’s group at Stanford University reported the discovery of 18 plasma proteins that could distinguish AD patients from non-demented controls with 90 percent accuracy (ARF related news story on Ray et al., 2007). Subsequent studies could not reproduce the findings (e.g., Soares et al., 2009). From the list of 18 plasma proteins, only interleukin-3 came up among the 17 analytes subsequently found to be associated with AD/MCI in the WashU and UPenn cohorts of the recent study by Hu et al., and they showed an opposite direction of association.

Wyss-Coray thinks the AD plasma biomarkers field may be at a similar stage as AD genetics was about 10 years ago. “Geneticists produced lists of more than 100 genes with linkage to AD, of which most did not hold up in much larger genomewide association studies,” Wyss-Coray said. “This showed that sample size is key. However, even if thousands of blood samples are analyzed, we still have the problem that protein assays and sample collection are not standardized.”

On a more hopeful note, the Stanford researchers developed a new approach using plasma measurements to predict pathological parameters in AD, and showed that the six proteins from their original 18-protein signature that were measured on the new platform could be replicated in an independent set of samples (Britschgi et al., 2011). “Most consistently we found changes in macrophage colony-stimulating factor (MCSF), granulocyte colony-stimulating factor (GCSF), and interleukin-3 (IL-3).”

Wyss-Coray and colleagues continue to use protein screens, measuring 600 or more proteins in plasma or CSF with antibody-based microarrays. “We have identified several interesting new proteins and pathways which we are now validating in biological assays and animal models of AD,” he said. “I think we will have to do this the hard way and link biology to any of the proteins that come out of screens before they are worth the effort to produce a clinical-grade assay.”—Esther Landhuis

Comments

1. • Tony Wyss-Coray
     Stanford University Medical School
   • Posted: 04 Sep 2012

   Our article (Ray et al., 2007) gained a lot of attention, but it was very early days and we had to work with what was available. Our samples were from multiple centers, and the cases and controls were not perfectly matched for each. There was also a difference in age between cases and controls, and the analytical platform we had used was a somewhat moving target, because the manufacturer (RayBiotech) made several changes to the array during the time we used it. Nevertheless, I think several of the markers we identified have biological relevance in AD and brain aging, and we are pursuing some of them successfully (e.g., MCSF). I would also draw attention to work from our lab that has been overlooked (Britschgi et al., 2011). We used an independent set of samples, a different analytical platform, and an innovative new approach to predict pathological parameters in AD using plasma markers as variables. Several models we developed reproduced six proteins out of the 18-protein Ray signature. Most consistently, we found changes in MCSF, GCSF and IL-3.

   Still, I think even now there are major challenges to find markers that will hold up in multiple studies across different centers and become clinical tools. It took maybe 10 years to achieve clinical utility with CSF Aβ and tau ELISAs, and I think it will take as long with any other protein-based assay (one at a time).

   The main problem is that protein measurements are extremely difficult to standardize, and multiplex assays are notoriously inexact. Major problems with current assays are that the reagents (antibodies, standards) are "research use only" and not clinical grade. They may, therefore, change from batch to batch, leading to variations in sensitivity and absolute concentrations for a given protein. Another, more trivial problem is that assays (e.g., ELISAs, Luminex) from different manufacturers may detect different isoforms of the same protein, active versus pre-proteins, or post-translationally modified proteins versus unmodified, leading sometimes to completely opposite results between groups.

   I think we are at a similar stage in this field as genetics was with SNP studies 10 years ago. Geneticists produced lists of more than 100 genes with linkage to AD, of which most did not hold up in the much larger GWAS. This showed that sample size is key. However, even if thousands of blood samples will be analyzed, we will still have the problem that the protein assays and sample collection are not standardized.

   Our lab continues to develop and use protein screens, and we currently measure more than 600 proteins in blood plasma or CSF using antibody-based microarrays. We know these arrays produce false-positive and -negative results, but we run several hundred samples in one batch to reduce variability. We have identified several interesting new proteins and pathways that we are now validating in biological assays and animal models of AD. I think we will have to go this hard way and link biology to any of the proteins that come out of screens before they are worth the effort to produce a clinical-grade assay.

   References:

   Ray S, Britschgi M, Herbert C, Takeda-Uchimura Y, Boxer A, Blennow K, Friedman LF, Galasko DR, Jutel M, Karydas A, Kaye JA, Leszek J, Miller BL, Minthon L, Quinn JF, Rabinovici GD, Robinson WH, Sabbagh MN, So YT, Sparks DL, Tabaton M, Tinklenberg J, Yesavage JA, Tibshirani R, Wyss-Coray T. Classification and prediction of clinical Alzheimer's diagnosis based on plasma signaling proteins. Nat Med. 2007 Nov;13(11):1359-62. PubMed.

   Britschgi M, Rufibach K, Huang SL, Clark CM, Kaye JA, Li G, Peskind ER, Quinn JF, Galasko DR, Wyss-Coray T. Modeling of pathological traits in Alzheimer's disease based on systemic extracellular signaling proteome. Mol Cell Proteomics. 2011 Oct;10(10):M111.008862. PubMed.

   View all comments by Tony Wyss-Coray

Make a Comment

To make a comment you must login or register.

References

News Citations

1. A Blood Test for AD? 15 Oct 2007

Paper Citations

1. O'Bryant SE, Xiao G, Barber R, Huebinger R, Wilhelmsen K, Edwards M, Graff-Radford N, Doody R, Diaz-Arrastia R, . A blood-based screening tool for Alzheimer's disease that spans serum and plasma: findings from TARC and ADNI. PLoS One. 2011;6(12):e28092. PubMed.
2. Soares HD, Potter WZ, Pickering E, Kuhn M, Immermann FW, Shera DM, Ferm M, Dean RA, Simon AJ, Swenson F, Siuciak JA, Kaplow J, Thambisetty M, Zagouras P, Koroshetz WJ, Wan HI, Trojanowski JQ, Shaw LM, . Plasma Biomarkers Associated With the Apolipoprotein E Genotype and Alzheimer Disease. Arch Neurol. 2012 Jul 16;:1-8. PubMed.
3. Johnstone D, Milward EA, Berretta R, Moscato P, . Multivariate protein signatures of pre-clinical Alzheimer's disease in the Alzheimer's disease neuroimaging initiative (ADNI) plasma proteome dataset. PLoS One. 2012;7(4):e34341. PubMed.
4. Koc M, Karaarslan O, Abali G, Batur MK. Variation in high-sensitivity C-reactive protein levels over 24 hours in patients with stable coronary artery disease. Tex Heart Inst J. 2010;37(1):42-8. PubMed.
5. Schmidt R, Schmidt H, Curb JD, Masaki K, White LR, Launer LJ. Early inflammation and dementia: a 25-year follow-up of the Honolulu-Asia Aging Study. Ann Neurol. 2002 Aug;52(2):168-74. PubMed.
6. Marioni RE, Stewart MC, Murray GD, Deary IJ, Fowkes FG, Lowe GD, Rumley A, Price JF. Peripheral levels of fibrinogen, C-reactive protein, and plasma viscosity predict future cognitive decline in individuals without dementia. Psychosom Med. 2009 Oct;71(8):901-6. PubMed.
7. Ray S, Britschgi M, Herbert C, Takeda-Uchimura Y, Boxer A, Blennow K, Friedman LF, Galasko DR, Jutel M, Karydas A, Kaye JA, Leszek J, Miller BL, Minthon L, Quinn JF, Rabinovici GD, Robinson WH, Sabbagh MN, So YT, Sparks DL, Tabaton M, Tinklenberg J, Yesavage JA, Tibshirani R, Wyss-Coray T. Classification and prediction of clinical Alzheimer's diagnosis based on plasma signaling proteins. Nat Med. 2007 Nov;13(11):1359-62. PubMed.
8. Soares HD, Chen Y, Sabbagh M, Roher A, Rohrer A, Schrijvers E, Breteler M. Identifying early markers of Alzheimer's disease using quantitative multiplex proteomic immunoassay panels. Ann N Y Acad Sci. 2009 Oct;1180:56-67. PubMed.
9. Britschgi M, Rufibach K, Huang SL, Clark CM, Kaye JA, Li G, Peskind ER, Quinn JF, Galasko DR, Wyss-Coray T. Modeling of pathological traits in Alzheimer's disease based on systemic extracellular signaling proteome. Mol Cell Proteomics. 2011 Oct;10(10):M111.008862. PubMed.

External Citations

1. Alzheimer’s Disease Neuroimaging Initiative

Further Reading

Papers

1. Britschgi M, Rufibach K, Huang SL, Clark CM, Kaye JA, Li G, Peskind ER, Quinn JF, Galasko DR, Wyss-Coray T. Modeling of pathological traits in Alzheimer's disease based on systemic extracellular signaling proteome. Mol Cell Proteomics. 2011 Oct;10(10):M111.008862. PubMed.
2. Yaffe K, Weston A, Graff-Radford NR, Satterfield S, Simonsick EM, Younkin SG, Younkin LH, Kuller L, Ayonayon HN, Ding J, Harris TB. Association of plasma beta-amyloid level and cognitive reserve with subsequent cognitive decline. JAMA. 2011 Jan 19;305(3):261-6. PubMed.
3. Doecke JD, Laws SM, Faux NG, Wilson W, Burnham SC, Lam CP, Mondal A, Bedo J, Bush AI, Brown B, De Ruyck K, Ellis KA, Fowler C, Gupta VB, Head R, Macaulay SL, Pertile K, Rowe CC, Rembach A, Rodrigues M, Rumble R, Szoeke C, Taddei K, Taddei T, Trounson B, Ames D, Masters CL, Martins RN, . Blood-Based Protein Biomarkers for Diagnosis of Alzheimer Disease. Arch Neurol. 2012 Jul 16;:1-8. PubMed.
4. Soares HD, Potter WZ, Pickering E, Kuhn M, Immermann FW, Shera DM, Ferm M, Dean RA, Simon AJ, Swenson F, Siuciak JA, Kaplow J, Thambisetty M, Zagouras P, Koroshetz WJ, Wan HI, Trojanowski JQ, Shaw LM, . Plasma Biomarkers Associated With the Apolipoprotein E Genotype and Alzheimer Disease. Arch Neurol. 2012 Jul 16;:1-8. PubMed.
5. Johnstone D, Milward EA, Berretta R, Moscato P, . Multivariate protein signatures of pre-clinical Alzheimer's disease in the Alzheimer's disease neuroimaging initiative (ADNI) plasma proteome dataset. PLoS One. 2012;7(4):e34341. PubMed.
6. O'Bryant SE, Xiao G, Barber R, Huebinger R, Wilhelmsen K, Edwards M, Graff-Radford N, Doody R, Diaz-Arrastia R, . A blood-based screening tool for Alzheimer's disease that spans serum and plasma: findings from TARC and ADNI. PLoS One. 2011;6(12):e28092. PubMed.
7. Ray S, Britschgi M, Herbert C, Takeda-Uchimura Y, Boxer A, Blennow K, Friedman LF, Galasko DR, Jutel M, Karydas A, Kaye JA, Leszek J, Miller BL, Minthon L, Quinn JF, Rabinovici GD, Robinson WH, Sabbagh MN, So YT, Sparks DL, Tabaton M, Tinklenberg J, Yesavage JA, Tibshirani R, Wyss-Coray T. Classification and prediction of clinical Alzheimer's diagnosis based on plasma signaling proteins. Nat Med. 2007 Nov;13(11):1359-62. PubMed.
8. O'Bryant SE, Xiao G, Barber R, Reisch J, Doody R, Fairchild T, Adams P, Waring S, Diaz-Arrastia R, . A serum protein-based algorithm for the detection of Alzheimer disease. Arch Neurol. 2010 Sep;67(9):1077-81. PubMed.

News

1. In the Blood: What Can Plasma Aβ Tell Us About Alzheimer’s? 22 Jan 2011
2. A Blood Test for AD? 15 Oct 2007
3. Plasma Aβ Testing Beset by Questions of Assays, Biology, Timing 12 Jun 2008